1. Technical Field of the Invention
The present invention relates to a sound signal encoding apparatus for and a sound signal encoding method of encoding and transmitting a sound signal, and more particularly to a sound signal encoding apparatus for and a sound signal encoding method of encoding and transmitting a sound signal such as a music sound signal data in a manner that the sound signal is encoded at a relatively high quality and smoothly transmitted to other electrically operating units via computer network.
2. Description of the Related Art
There have so far been proposed a wide variety of sound signal encoding apparatuses of this type one typical example of which is shown in FIG. 20. The conventional sound signal encoding apparatus comprises sound signals dividing means 101, first sound signal sections analyzing means 102, sampling rate selecting means 103, and sound signal sampling means 104. The sound signals dividing means 101 is operative to divide each of two different sound signals into a plurality of sound signal sections along a time axis for each of the sound signals to be taken for receiving therein. The sound signals consist of a first channel signal and a second channel signal. The first sound signal sections analyzing means 102 is designed to analyze each of the divided sound signal sections based on the sound signal characteristics inherent in the sound signal. The sampling rate selecting means 103 is adapted to select one arbitrary sampling rate for each of the sound signal sections from among predetermined sampling rates. The sound signal sampling means 104 is operative to sample each of the analyzed sound signal sections at the sampling rate selected by the sampling rate selecting means 104.
The conventional sound signal encoding apparatus further comprises second sound signal sections analyzing means 105, frequency components calculating means 106, and quantization bit numbers allocating means 107. The second sound signal sections analyzing means 105 is operative to analyze a masking threshold level for each of the divided sound signal sections based on a psycho acoustic model obtained by taking advantage of human's hearing characteristics.
The frequency components calculating means 106 is operative to calculate frequency components with two different information consisting of first and second signals for each of the sound signal sections sampled at the selected sampling rate, the above first signal being indicative of intensities, while the above second signal being indicative of frequencies. The quantization bit numbers allocating means 107 is designed to allocate quantization bit numbers for each of the calculated frequency components for each of the sound signal sections.
The conventional sound signal encoding apparatus further comprises first frequency components compressing means 108, and second frequency components compressing means 109. The first frequency components compressing means 108 is adapted to compress the frequency components for each of the sound signal sections with two different information consisting of first and second signals. The second frequency components compressing means 109 is operative to compress the frequency components for each of the sound signal sections with two different information consisting of first and second signals. The above first signal is indicative of the addition of each of the frequency components for the first channel signal and each of the frequency components for the second channel signal, while the above second signal is indicative of the difference between each of the frequency components for the first channel signal and each of the frequency components for the second channel signal.
The conventional sound signal encoding apparatus further comprises frequency components quantizing means 110 and frequency components encoding means 111. The frequency components quantizing means 110 is operative to quantize each of the frequency components for each of the sound signal sections at the predetermined quantization bit numbers. The frequency components encoding means 111 is operative to encode the quantized frequency components for each of the sound signal sections to a multiplexed bit stream. The multiplexed bit stream is constituted by the sound signals for each of the sound signal sections and general information needed for the sound signals to be encoded and decoded.
The sound signal is encoded by the conventional sound signal encoding apparatus in accordance with the MPEG2 AAC (Advanced Audio Coding) decided by the Motion Picture Experts Group, and is then transmitted at a predetermined transmitting bit rate to other electrically operating units via computer network.
The conventional sound signal encoding apparatus thus constructed in the above encounters such a problem that the sound signal tends to be encoded at a relatively low quality. The reason is due to the fact that the first frequency components compressing means 108 is operative to compress the frequency components for each of the sound signal sections with two different information consisting of the first and second signals. The above second signal is intended to indicate the intensity ratio of one of the frequency components for the first channel signal and the frequency components for the second channel signal to the other of the frequency components for the first channel signal and the frequency components for the second channel signal. The compression of the sound signal thus performed by the first frequency components compressing means 108 results in an excessive compression to the sound signal, and contributing to a wasteful load to computers building the network and deteriorating a music sound quality when the sound signal is decoded.